53

ECOM II

Click OK to run the analysis and display the results on the Regression tab.

With backward stepwise regression, because variables are removed, rather than being

added there is no minimum probability for removal and thus the probability boxes are

not displayed.

Click OK to run the analysis and display the results on the Regression tab.

9.2

Foward Stepwise Linear Regression

Forward stepwise regression starts with no variables in the model. It then adds the

most significant explanatory variable, that is, the one with the lowest p value, at each

step, until all variables have been added. By scrutinising the overall fit of the model

variables will be automatically added (and, if they do not improve the fit, removed

again) until the optimum model is found.

Load both your species and your environmental datasets as described in

Opening a data set

From the Regression menu on the Toolbar select Forward Stepwise. A list of

dependent variables (usually species) is displayed.

The results report shows the sequences of the procedure as Steps:

Copyright 2004  PISCES Conservation Ltd